Coding Conventions

Core Sections

Item	Rule	Example	Evidence
Files	snake_case.py	matching_engine.py, order_book.py, executed_trades.py	src/order_matching/ directory
Functions/methods	snake_case, descriptive verb phrases	`match()`, `get_imbalance()`, `_execute_trades()`	matching_engine.py, order_book.py
Classes	PascalCase	`MatchingEngine`, `OrderBook`, `LimitOrder`, `ExecutedTrades`	All source files
Private methods	Leading underscore: `_method_name`	`_match()`, `_execute_trade()`, `_get_expired_orders()`	matching_engine.py lines 74-134
Constants/enum values	UPPER_SNAKE_CASE for enum values	`Side.BUY`, `Side.SELL`, `Status.OPEN`, `Status.CANCEL`	side.py, status.py
Type aliases	PascalCase with `Type` suffix	`OrderBookOrdersType = dict[float, Orders]`	order_book.py line 13

Formatter: ruff-format (.pre-commit-config.yaml line 21-23)
Linter: ruff (.pre-commit-config.yaml line 17-20, pyproject.toml lines 66-83)
Most relevant enforced rules:
Line length: 120 characters (pyproject.toml line 67)
Select: E (pycodestyle errors), F (pyflakes), D (pydocstyle), B (bugbear), I (isort), ARG (unused arguments), ANN (type annotations) (pyproject.toml line 71)
Docstring convention: NumPy style (pyproject.toml line 86)
Ignore: D100-D107 (missing docstrings for modules/classes/methods, except public functions), D213, D417 (pyproject.toml lines 72-83)
Run commands: ```bash # Run linter with auto-fix uv run ruff check --fix

# Run formatter uv run ruff format

# Run via pre-commit (includes ty type-check) uv run prek run -v --show-diff-on-failure --all-files ```

Import grouping/order: Enforced by ruff isort (I rules). Standard library, third-party, local imports separated by blank lines (pyproject.toml line 71)
Alias vs relative import policy: Absolute imports from order_matching.* used throughout. No relative imports detected. Examples: from order_matching.order import Order, from order_matching.side import Side
Public exports/barrel policy: No __all__ exports defined. Main module __init__.py only contains version detection (src/order_matching/init.py)

Error strategy by layer: Assertions used for internal state validation (e.g., assert self._timestamp is not None in matching_engine.py line 76, line 121). No custom exception types defined. Type validation handled by dataclasses and type hints.
Logging style and required context fields: No logging library detected in codebase. Library pattern—users are responsible for logging in their applications.
Sensitive-data redaction rules: Not applicable (no PII or secrets in order matching library)

Test file naming/location rule: test_*.py files in tests/ directory, separate from source. Examples: test_matching_engine.py, test_order_book.py, test_orders.py (tests/ directory listing)
Mocking strategy norm: Use pytest fixtures for test data generation. Faker with fixed seeds for reproducible random data (tests/conftest.py). No external service mocking needed (library has no I/O).
Coverage expectation: Coverage measured with pytest-cov, run in CI with --cov=order_matching flag (.github/workflows/workflow.yaml line 34). No explicit threshold set in config. Doctests enabled for README.md and module docstrings (pyproject.toml line 63).